From our ivory tower in the kingdom of blogdom, we track cases and litigations from afar, peeking in on them from decisions rendered at specific points in time. Sometimes, from a single decision, we venture on what will happen next, like whether claims that survived dismissal will make it past summary judgment or whether a judgment will survive appeal. Other times, we look at multiple decisions in the same case or litigation and make a somewhat more intelligent guess as to where things are headed. Every once in a while, we have looked at several pre-trial rulings from the same case that consistently favor one side or position that it is fairly easy to guess what the judge will do at trial. What the jury will do is always less predictable.
In In re Actos (Pioglitazone) Prods. Liab. Litig. (Allen v. Takeda Pharms. N. Am., Inc.), MDL No. 6:11-md-2299, No. 12-cv-000064-RFD-PJH, 2014 U.S. Dist. LEXIS 121648 (W.D. La. Aug. 28, 2014), we have the culmination of several bad decisions. Back in January, we questioned the court’s rejection of conflict preemption for failure to warn claims in the absence of evidence that the FDA had rejected a request to strengthen a warning of an approved drug. We also suggested that the court’s requirement of only a prima facie showing by the plaintiff allowed it five experts to stroll through the Daubert gate without any real challenge by its purported keeper. Later, we have posted on rulings on spoliation and a request for default judgment because a witness answered “I don’t know” a number of times. Along the way, we noted that the first MDL bellwether trial had resulted in a very high plaintiff verdict with lots of punitive damages thrown in. After several months, the court has now ruled on the defendants’ Rule 50(b) motion made during trial and re-urged after the verdict. In a surprise to nobody following the case, the court denied the motion in its entirety. And nobody who follows this blog will be surprised to hear that we think that the decision was wrong and that the reasoning of the opinion was lacking.
We start by what the reader will not find in the opinion—what the jury awarded to the plaintiff. That was $1.5 million in compensatory damages and $9 billion (with a “b”) in punitive damages. While there is plenty of discussion of money in the opinion, the profits from the sale of the drug and how profits allegedly motivated the NDA holder and co-marketer to take various actions concerning the drug and its bladder cancer risk, but nothing about what the jury awarded, whether it bore any relation to the evidence, or whether it provided any basis to suspect that something had gone wrong with the trial. While the opinion did not address any challenge to the amount of the punitive award—just to whether there was sufficient evidence to support any award (particularly against a mere co-promoter)—it does seem strange to omit the amount of the award from a very long decision. It also seems weird that the court repeatedly expressed its misgivings about having to address issues it had addressed in different procedural postures when, as it said, the “jurisprudential support” had not changed. This is, of course, how the Federal Rules of Civil Procedure work. A motion to dismiss may be followed by a motion for summary judgment, which may be followed by motions for judgment as a matter of law during and after trial, all of which may address the same issue (e.g., preemption) without the controlling law changing. Much like the analysis of whether the amount of punitive damages comports with constitutional requirements involves looking at proportionality in regard to the amount of actual damages, there is some notion that very jury large awards, jury decisions involving many plaintiffs, and even defense verdicts in cases where the plaintiff had substantial injuries should receive extra scrutiny from the presiding judge. An opinion that conveys a sense of anger at some of the parties (or their lawyers) or a reluctance to consider all the arguments at face value does not increase confidence in our civil courts.
As we have said, the opinion is quite long, in part because it recaps the plaintiff’s version of the evidence he offered in more than two months of trial. We are not in a position to begin to question whether the evidence was as described in the opinion, but a few things are obvious from reviewing the recap and how it is used. First, lots of evidence came in that was focused solely on whether defendants defrauded the FDA. Second, lots of evidence came in that blurred the lines between which defendant did what and when they did it in relation to when plaintiff was prescribed the drug. Third, although Dr. Kessler was plaintiff’s mouthpiece for why the label should have been changed earlier and how discussions with FDA on labeling should have been different, there does not appear to have been any evidence that the label could have been changed unilaterally or that FDA would have acceded to the labeling change plaintiff urged. Fourth, in discussions about what the co-marketer’s sales representatives did not say, there was no attention to whether the additional information that plaintiff urged could have been legally disclosed before the label had changed. These observations tie in to the court’s ultimate rejection of the defendants’ preemption and punitive damages arguments. There is certainly more to the opinion, but we can only address so much in one post before the Oscar wrap-it-up music starts playing in our head.
We would be remiss if we discussed the rejection of defendants’ preemption and punitive damages challenges without noting that the opinion does not mention Buckman, which held:
[T]he plaintiffs’ state-law fraud-on-the-FDA claims conflict with, and are therefore impliedly pre-empted by federal law. The conflict stems from the fact that the federal statutory scheme amply empowers the FDA to punish and deter fraud against the Agency, and that this authority is used by the Agency to achieve a somewhat delicate balance of statutory objectives. The balance sought by the Agency can be skewed by allowing fraud-on-the-FDA claims under state tort law.
531 U.S. 341, 348 (2001). This ruling was not based on the express preemption provisions of the Medical Device Amendment and has been applied to drug cases fairly often. See, e.g., Garcia v. Wyeth-Ayerst Labs., 385 F.3d 961, 965-966 (6th Cir. 2004); Bouchard v. American Home Prods. Corp., 213 F. Supp. 2d 802, 811 (N.D. Ohio 2002). Even from an account of the evidence that is avowedly from plaintiff’s perspective, the plaintiff’s case was about as focused on whether defendants had defrauded the FDA as any we have seen since Buckman. Deciding issues of preemption and punitive damages without addressing whether plaintiff was advancing fraud-on-the-FDA claims, however he titled them, makes no sense to us. Indeed, the opinion’s repeated citation of plaintiff’s evidence that allegedly showed defendants had concealed relevant information from the FDA as a justification for why there was no preemption of claims or why plaintiff could recover punitive damages against both defendants just reinforces that these were fraud-on-the-FDA claims that should have been preempted.
As we have said many times, a conflict preemption analysis should start with identifying whether there is a properly asserted and supported claim under a recognized state law theory of recovery and what the regulated defendant needed to do to comply with the state law duty. Then, the analysis should go on to whether the defendant can comply with the state law duty while simultaneously complying with its federal duties. (There are many in-depth discussions of preemption principles in prior posts, including this one on Bartlett, another Supreme Court preemption decision not discussed in the opinion.) In analyzing whether the claims asserted against the co-marketer were impliedly preempted, the opinion neither cites the state law providing for such a claim nor what plaintiff contended the co-marketer was supposed to have do meet its state law duties. 2014 U.S. Dist. LEXIS 121648, **69-79. In fairness—we do want to be fair sometimes—there are cites in two other sections to a single intermediate appellate decision “that visits liability on companies . . . that are responsible for placing a drug into the marketplace.” Id. at **59 & 175 (citing Brumbaugh v. CEJJ, Inc., 152 A.D.2d 69, 71, 547 N.W.S.2d 699 (N.Y. Sup. Ct. – App. Div. 3d Dept. 1989)). That is not the same thing as saying the state law imposes duties on co-marketers, the breach thereof can result in liability for failure to warn and implied warranty of merchantability, the two claims plaintiff apparently advanced at trial. Similarly, for all the citation to plaintiff’s evidence that the co-marketer did not disclose the true risk of bladder cancer in the years before the FDA-approved label did (from plaintiff’s perspective), there is nothing in the opinion saying what an adequate warning would have been. This matters because the opinion’s analysis of preemption boiled down to whether the co-marketer was precluded by federal law from conveying to physicians “any information or warning language beyond the actual language included on the insert label approved by the FDA.” Id. at *70. Concluding that the regulations afforded some leeway to present marketing materials that were different than, but still consistent with, the approved label, and pointing to evidence that the co-marketer consulted with the NDA holder on the label and did have some marketing materials that were different than the label, the court found no preemption. This is not the right inquiry. Did state law impose a duty on the co-marketer to provide physicians with information about the risk of the product? If it did, then could the co-marketer have complied with federal requirements of keeping marketing materials and statements consistent with the approved label while including whatever extra-label statements about a bladder cancer risk that the plaintiff urged? Those are the questions that should have been answered.
The opinion’s consideration of the broader preemption challenge made by both defendants followed a similar pattern. The issue was whether FDA would have approved the labeling change that plaintiff urged—whatever that actually was—before plaintiff allegedly developed bladder cancer from using the drug. This would be evaluated under the singular “clear evidence” standard from Levine, but with a free hand for plaintiff to ignore Buckman and contend that defendants kept FDA in the dark on the risk of bladder cancer. Defendants had evidence that FDA determined, in the year plaintiff started the drug and the co-marketer stopped co-marketing it, that a bladder cancer warning belonged in the Precautions section of the label, did not request a revision upon receipt of study data three years later, and only approved a label with information on bladder cancer in the Warnings section two years after that. Plaintiff countered with evidence that the defendants allegedly withheld information from the FDA and that defendants negotiated with FDA over the label, but nothing about whether the FDA would have approved the warning plaintiff wanted when plaintiff wanted them.
In weighing this evidence, the court seemed to endorse the ludicrous notion, advanced by Dr. Kessler, that language in a drug label about a risk of the drug does not count as a “warning” (a state law concept) unless it is in the “Warning” section of the label. Id. at **47-48; but see id. at *168 (“The Defendants are, however, correct that New York law does not require a warning to be located in the “Warnings” section of a medication label in order for the warning to comply with the duty to warn individuals considering taking the medication, or physicians considering prescribing the medication. However, [plaintiff says none of the information about bladder cancer in the label should be considered a warning].”). Setting aside how the format of drug labels changed during the time at issue in this case due to the Physician Labeling Rule so that the old “Warnings” and “Precautions” sections have been merged into one section called “Warnings and Precautions,” this is just a naïve view prescription drugs. Language in other sections of drug labeling, like Contraindications, may be key to individual prescribing decisions and the focus of allegations by product liability plaintiffs. In addition to this misimpression, the opinion ignored the possibility that the FDA would ever find a proposed warning to be excessive in light of the scientific evidence available.
Finally, in light of the evidence pointed to by Plaintiffs as presented at trial, this Court cannot find Defend-ants’ argument that the FDA, had it been presented with a complete, accurate, and forthright description of the evidence, would have chosen to hide from the medical community and the general public the possibility of an increased risk of the very serious side effect of bladder cancer by not allowing the very warning they made overture to explore, persuasive.
Id. at *85 (emphasis in original). We could go on, but we think you get the gist.
The rejection of all arguments as to the punitive award (amount hidden from the reader) built on the evidence of fraud-on-the-FDA and of the defendants failure to do things they may not have been permitted to do by applicable regulations. It also looked to evidence of what you read as general alleged “bad conduct” evidence not tied to the underlying theories of recovery, which is generally frowned upon when it comes to punitive damages. On top of this morass of evidence that the court stated (several times) showed that the defendants intentionally picked profits over safety, the jury got to hear about the alleged spoliation of records by the NDA holder, which we have already said mostly involved missing from sales representatives who left the company five or more years before the litigation started. The NDA holder’s “conduct in destroying files of key employees involved in the development, marketing, and management of Actos® could be one fact the jury might have considered in assessing Takeda’s intent.” Id. at *179 (emphasis in original). The jury was also, apparently, permitted to infer something about the co-marketer’s intent from this “fact.” There was nothing wrong with that, because the co-marketer had apparently not requested a limiting instruction when the jury was informed of court’s view on spoliation. Id. at **180-81. The court took no responsibility for informing the jury of spoliation by one defendant without making it clear that the other defendant was not implicated.
We will have to wait and see if the court can find some way under Rule 59 (motion still pending) to justify a punitive damages award that was 6000 times higher than the actual damages award. We will also have to wait and see what, if anything, the Fifth Circuit has to say about all of this. We do know its decision in Lofton v. McNeil Consumer & Specialty Pharms., 672 F.3d 372 (5th Cir. 2012), embraced Buckman, so we are optimistic that a panel of cooler heads would prevail.